CROI 2013: VOICE Trial Results on Daily HIV Prevention for Women
Highly anticipated results were reported today from the VOICE trial, which looked at the safety and efficacy of daily oral PrEP and drug-containing vaginal microbicide gel in more than 5,000 women in South Africa, Uganda, and Zimbabwe.
Jeanne Marrazzo, MD, MPH, explained to a packed auditorium at the 20th Retrovirus Conference that these approaches did not prevent new HIV infections in this particular study because most participants didn’t actually use them.
When VOICE—short for Vaginal and Oral Interventions to Control the Epidemic—began enrolling women in September 2009, it had five study groups. Participants were randomized to use one of the following products daily:
- tenofovir gel
- placebo gel
- oral tenofovir tablet
- oral Truvada (the tenofovir/emtricitabine combination)
- oral placebo pill
(The trial also included the standard HIV prevention “package”: free condoms, HIV testing and safer sex counseling, and screening and treatment for other sexually transmitted infections. Adherence to condom use was self-reported as 85% overall.)
In the fall of 2011, the study arms testing oral tenofovir and tenofovir gel were stopped when it became clear that although both were proving to be safe, neither product was working to prevent HIV infection better than the corresponding placebo products. Evaluation of the Truvada arm continued until the scheduled study end date in August 2012.
To assess whether women were adhering to their product regimen, researchers collected unused pills and gel at study visits, and participants self-reported their adherence in audio computer-assisted self-interviews (known as A-CASI). Based on these measures, researchers estimated adherence to be roughly 90% overall.
But this high level of adherence jarred with the 312 new HIV infections that occurred during the study. Marrazzo reported today that overall HIV incidence (the rate of new infections) in VOICE was 5.7%, with no statistically significant differences in HIV incidence between the active drug arms and the matched placebo groups. (For a product-by-product breakdown of HIV incidence, see the “Understanding the Results of VOICE” fact sheet released today by the Microbicide Trials Network.)
Further contradicting these estimates of adherence, an analysis of 3,298 blood samples taken at quarterly visits from 773 participants revealed that 58% of those in the oral tenofovir arm had no detectable tenofovir in any sample. A similar pattern emerged for oral Truvada recipients and tenofovir gel recipients: 50% and 55%, respectively, had no active drug in any of their blood samples. (But unfortunately, said Marrazzo, those data do not translate into a significantly lower HIV-acquisition risk for women with detectable tenofovir.)
So who did take the study products? Women who were older than 25 years, women who were married, and women whose primary male partner was 28 years old or older had higher likelihoods of tenofovir detection in their blood. Judging by these blood sample data, adherence was especially low “among younger, unmarried women who were most at risk for HIV.”
The vast disconnect seen in this trial between self-reported adherence to the study products and the actual detection of study drugs in participants’ blood illuminates the potential unreliability of self-reported data and has implications for the design of future trials. “It casts skepticism on the validity of any self-report,” Marrazzo said, “from sexual activity to condom use to what you had for dinner last night.”
Women may have been reluctant to report their low adherence and put their continued participation in jeopardy, she speculated, because the benefits of participating in the trial were so highly valued. These benefits included social interaction with study staff and other participants, as well as clinical services like Pap tests and screenings for sexually transmitted infections.
The VOICE findings, Marrazzo added, “clearly support the pursuit of alternate delivery of agents that are not dependent on daily use.” In a press conference, she discussed the possibility of pursuing a longer-acting product—that is, if women want a product specifically designed to prevent HIV infection. Stigma around using an HIV drug may have been a key factor in the low adherence seen in VOICE, she observed.
To help get around this barrier, future trials of PrEP for women might include “stealth delivery” of anti-HIV agents in combination with a contraceptive, suggested press conference chair Sharon Hillier, PhD. Marrazzo added that options for intermittent dosing should also be explored for women who might prefer to use a product only with sexual activity rather than long term.
“The bottom line,” she concluded, “is that no one size fits all.”
Reilly O’Neal is a freelance writer and former editor of BETA.
Marrazzo, J. and others. Pre-exposure prophylaxis for HIV in women: daily oral tenofovir, oral tenofovir/emtricitabine, or vaginal tenofovir gel in the VOICE study (MTN 003). 20th Conference on Retroviruses and Opportunistic Infections. Atlanta. March 3–6, 2013. Abstract #26LB.
Microbicide Trials Network. Daily HIV prevention approaches didn’t work for African women in the VOICE study. Press release. March 4, 2013.